2005-09-10 23:49:46

by Andrew Smith

[permalink] [raw]
Subject: kernel performance problem

This is my first ever posting to this list, so please excuse me if I've
f**cked up and broken protocol.
I use an IDE LG DVD-RW drive for backup(GSA-4081B).
Back around 2.6.10 it wrote at typically 1.6-1.7x speed, but thereafter
it dropped to around 0.6-0.7 x.


%hdparm -I /dev/hdd

/dev/hdd:

ATAPI CD-ROM, with removable media
Model Number: HL-DT-ST DVDRAM GSA-4081B
Serial Number: K2B3CFE4600
Firmware Revision: A100
Standards:
Likely used CD-ROM ATAPI-1
Configuration:
DRQ response: 50us.
Packet size: 12 bytes
Capabilities:
LBA, IORDY(can be disabled)
DMA: mdma0 mdma1 mdma2 udma0 udma1 *udma2
Cycle time: min=120ns recommended=120ns
PIO: pio0 pio1 pio2 pio3 pio4
Cycle time: no flow control=120ns IORDY flow control=120ns
HW reset results:
CBLID- below Vih
Device num = 1


--
Andrew




2005-09-11 12:30:07

by Walt H

[permalink] [raw]
Subject: Re: kernel performance problem

>
>
>This is my first ever posting to this list, so please excuse me if I've
>f**cked up and broken protocol.
>I use an IDE LG DVD-RW drive for backup(GSA-4081B).
>Back around 2.6.10 it wrote at typically 1.6-1.7x speed, but thereafter
>it dropped to around 0.6-0.7 x.
>
>
>%hdparm -I /dev/hdd
>
>/dev/hdd:
>
>ATAPI CD-ROM, with removable media
> Model Number: HL-DT-ST DVDRAM GSA-4081B
> Serial Number: K2B3CFE4600
> Firmware Revision: A100
>Standards:
> Likely used CD-ROM ATAPI-1
>Configuration:
> DRQ response: 50us.
> Packet size: 12 bytes
>Capabilities:
> LBA, IORDY(can be disabled)
> DMA: mdma0 mdma1 mdma2 udma0 udma1 *udma2
> Cycle time: min=120ns recommended=120ns
> PIO: pio0 pio1 pio2 pio3 pio4
> Cycle time: no flow control=120ns IORDY flow control=120ns
>HW reset results:
> CBLID- below Vih
> Device num = 1
>

Hi Andrew,

I know that hdparm shows udma2 mode selected for data transfers, but
check that dma is *actually* turned on. ie: hdparm -d1 /dev/hdd

I don't remember exactly when this went in, but there were some IDE
patches around the 2.6.11 or so timeframe that seemed to change the way
these devices get setup at boot. I believe this happened when the
autotune= syntax was deprecated. HTH,

-Walt